Leon/docutranslate
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

small fix

Browse Source
This commit is contained in:
xunbu
2025-05-11 12:06:28 +08:00
parent 6f3b02f562
commit 2bb6e68144
2 changed files with 5 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
.idea/workspace.xml generated
View File

@@ -5,12 +5,7 @@
</component> </component>
<component name="ChangeListManager"> <component name="ChangeListManager">
<list default="true" id="6b18b44a-df57-4212-a857-9e291ebe5dd2" name="更改" comment=""> <list default="true" id="6b18b44a-df57-4212-a857-9e291ebe5dd2" name="更改" comment="">
<change beforePath="$PROJECT_DIR$/.idea/workspace.xml" beforeDir="false" afterPath="$PROJECT_DIR$/.idea/workspace.xml" afterDir="false" /> <change beforePath="$PROJECT_DIR$/docutranslate/utils/convert.py" beforeDir="false" afterPath="$PROJECT_DIR$/docutranslate/utils/convert.py" afterDir="false" />
<change beforePath="$PROJECT_DIR$/README.md" beforeDir="false" afterPath="$PROJECT_DIR$/README.md" afterDir="false" />
<change beforePath="$PROJECT_DIR$/docutranslate/Agents/__init__.py" beforeDir="false" afterPath="$PROJECT_DIR$/docutranslate/agents/__init__.py" afterDir="false" />
<change beforePath="$PROJECT_DIR$/docutranslate/Agents/agent.py" beforeDir="false" afterPath="$PROJECT_DIR$/docutranslate/agents/agent.py" afterDir="false" />
<change beforePath="$PROJECT_DIR$/docutranslate/Agents/markdown_agent.py" beforeDir="false" afterPath="$PROJECT_DIR$/docutranslate/agents/markdown_agent.py" afterDir="false" />
<change beforePath="$PROJECT_DIR$/docutranslate/translater.py" beforeDir="false" afterPath="$PROJECT_DIR$/docutranslate/translater.py" afterDir="false" />
</list> </list>
<option name="SHOW_DIALOG" value="false" /> <option name="SHOW_DIALOG" value="false" />
<option name="HIGHLIGHT_CONFLICTS" value="true" /> <option name="HIGHLIGHT_CONFLICTS" value="true" />
@@ -371,7 +366,7 @@
<workItem from="1746887311759" duration="14000" /> <workItem from="1746887311759" duration="14000" />
<workItem from="1746894355924" duration="651000" /> <workItem from="1746894355924" duration="651000" />
<workItem from="1746923054682" duration="383000" /> <workItem from="1746923054682" duration="383000" />
<workItem from="1746928498783" duration="4163000" /> <workItem from="1746928498783" duration="4523000" />
</task> </task>
<servers /> <servers />
</component> </component>
@@ -382,7 +377,7 @@
<SUITE FILE_PATH="coverage/filetranslate$test.coverage" NAME="test 覆盖结果" MODIFIED="1746677277745" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" /> <SUITE FILE_PATH="coverage/filetranslate$test.coverage" NAME="test 覆盖结果" MODIFIED="1746677277745" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
<SUITE FILE_PATH="coverage/filetranslate$agent_utils.coverage" NAME="agent_utils 覆盖结果" MODIFIED="1746708534311" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate/utils" /> <SUITE FILE_PATH="coverage/filetranslate$agent_utils.coverage" NAME="agent_utils 覆盖结果" MODIFIED="1746708534311" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate/utils" />
<SUITE FILE_PATH="coverage/filetranslate$convert.coverage" NAME="convert 覆盖结果" MODIFIED="1746780691113" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate/utils" /> <SUITE FILE_PATH="coverage/filetranslate$convert.coverage" NAME="convert 覆盖结果" MODIFIED="1746780691113" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate/utils" />
<SUITE FILE_PATH="coverage/filetranslate$test1.coverage" NAME="test1 覆盖结果" MODIFIED="1746935919416" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" /> <SUITE FILE_PATH="coverage/filetranslate$test1.coverage" NAME="test1 覆盖结果" MODIFIED="1746936018440" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
<SUITE FILE_PATH="coverage/PDFtranslate$PDFtranslater__1_.coverage" NAME="PDFtranslater (1) 覆盖结果" MODIFIED="1746633258205" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/pdftranslate_packages" /> <SUITE FILE_PATH="coverage/PDFtranslate$PDFtranslater__1_.coverage" NAME="PDFtranslater (1) 覆盖结果" MODIFIED="1746633258205" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/pdftranslate_packages" />
<SUITE FILE_PATH="coverage/PDFtranslate$convert.coverage" NAME="convert 覆盖结果" MODIFIED="1746596984213" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/pdftranslate_packages/utils" /> <SUITE FILE_PATH="coverage/PDFtranslate$convert.coverage" NAME="convert 覆盖结果" MODIFIED="1746596984213" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/pdftranslate_packages/utils" />
<SUITE FILE_PATH="coverage/PDFtranslate$agent_utils.coverage" NAME="agent_utils 覆盖结果" MODIFIED="1746617703678" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/pdftranslate_packages/utils" /> <SUITE FILE_PATH="coverage/PDFtranslate$agent_utils.coverage" NAME="agent_utils 覆盖结果" MODIFIED="1746617703678" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/pdftranslate_packages/utils" />

4
docutranslate/utils/convert.py
View File

@@ -6,7 +6,7 @@ from docling_core.types.doc import ImageRefMode
from pathlib import Path from pathlib import Path
from docling.document_converter import DocumentConverter, PdfFormatOption from docling.document_converter import DocumentConverter, PdfFormatOption
from docling.datamodel.document import DocumentStream from docling.datamodel.document import DocumentStream
from docutranslate.logger import translater_logger
IMAGE_RESOLUTION_SCALE = 4 IMAGE_RESOLUTION_SCALE = 4
@@ -25,7 +25,7 @@ def file2markdown_embed_images(file_path: Path | str|DocumentStream, formula=Fal
try: try:
result = converter.convert(file_path).document.export_to_markdown(image_mode=ImageRefMode.EMBEDDED) result = converter.convert(file_path).document.export_to_markdown(image_mode=ImageRefMode.EMBEDDED)
except LocalEntryNotFoundError: except LocalEntryNotFoundError:
print(f"无法连接huggingface，正在尝试换源") translater_logger.info(f"无法连接huggingface，正在尝试换源")
os.environ['HF_ENDPOINT'] = 'https://hf-mirror.com' os.environ['HF_ENDPOINT'] = 'https://hf-mirror.com'
result = converter.convert(file_path).document.export_to_markdown(image_mode=ImageRefMode.EMBEDDED) result = converter.convert(file_path).document.export_to_markdown(image_mode=ImageRefMode.EMBEDDED)
return result return result